Presenter

Javier is a Senior Research Engineer at Barcelona Supercomputing Center (BSC). He is the Head of Data and Model Evaluation at the Language Technologies Lab. On the Data side, his and his teams’s work aims at acquiring pre- and post-training data in a legal and traceable way, at generating and validating synthetic data, and at efficiently processing the data and finding optimal data distributions for training. On the Evaluation side, the team aims at continue developing a comprehensive evaluation setup for multilingual language models encompassing capabilities, bias and safety, at generating high-quality evaluation multilingual datasets, and at improving the human side of model evaluation. Javier's current research lines and interests are connected with benchmarking efficiency, improving reliability and validity of LLM-as-a-judge, and evaluation methods for the biomedical and health domains. He is also part of the Trillion Parameter Consortium (TPC) Planning Team and Steering Group, a core member in the TPC Skill, Safety and Trust Evaluation working group, and a member of the Trust & Safety and Open Foundation Models & Datasets working groups of the AI Alliance. He holds a PhD in Psycholinguistics, an MSc in Natural Language Processing and an MSc in Language Education. He has taught and lectured in graduate and postgraduate courses in universities such as The University of Edinburgh, Universidad de Zaragoza, and Imperial College London. He has also worked as external quality assessor for Madri+d and AQU Catalunya, and is currently a member of the Evaluation and Certification Committee at Madri+d.